My own petard

I've been complaining along with the rest of the E-Rate community about the timing of audits. Why did they have to come right at the busiest time of the year? Why not have them start in March or April?

Now I'm getting my "be careful what you wish for" comeuppance: two notifications in the last week or so of upcoming audits. Actually, it is much better than getting all the requests at once, but I'm still trying to close up one of the earlier audits, so it feels like piling on.

I did a quick check, and looks like 100% of my clients with disbursements of over $500,000 in 2007-2008 got audited. It's not a big group, so maybe I'm just unlucky.

2-in-5 is 0-5

I'm a little late in commenting on this, since I'm facing a flurry of audits.

USAC released its demand estimate for 2009-2010, and once again, the "2-in-5 Rule" (which says that applicants can only purchase Internal Connections twice in a 5-year period) has failed. The stated purpose of the rule was to decrease demand, but it just doesn't work.

This year the Internal Connections demand among 90% schools is down significantly from last year, but still above 2007 levels. There was a big bump in demand in 2008, and now we have returned to the steady increase in demand which we've seen since the 2-in-5 Rule took effect.

I'm thinking that this year's denial threshold will be low (60%?) because USAC has a big pile of money sitting around ready to be rolled over, but I give the 2-in-5 Rule no credit for that.

I've said it before and I'll say it again (and again and again): the 2-in-5 Rule must go.

What's the protocol?

We've noticed one small change to the online Item 21 Attachment application, and I don't like it.

For FRNs in the Internet Access category, the drop-down menu of services used to have one choice for "Broadband Internet access." Now there are several items like "DSL" and "Frame relay" and "T-1."

I understand why the change was made. In order to approve an FRN for Internet access, PIA always wants:

1. Number of connections
2. Bandwidth (speed) of connection(s)
3. Type(s) of connection(s) (cable, T-1, etc.)

So it makes sense to have the Item 21 Attachment ask for all those things. Up until this year the form did not actually ask for any of that info. You were just supposed to put it in the "description" box.

I agree that the current method of collecting data makes it useless for any kind of analysis, so I suppose it's good that the form now asks for type of connection. Except it's not quite right. Here are the problems I have with it.

First, the items say "DSL" and "Frame relay" not "DSL Internet access." OK, anyone techie will figure out what's meant, but I guarantee at least one applicant is going to see "Frame relay" there and think that the frame relay circuits connecting their buildings should be in Internet Access, which is wrong. Even worse, this is going to feed the misunderstanding about the allowability of WANs under Internet Access, which is muddied enough already. The items should be "Internet access: DSL" and "Internet access: T-1" and so on. That way, the list would sort nicely, with all the Internet access items together.

Next problem: not all the possible technologies are there. Here in NJ, many districts are still using ATM, but that is not one of the options. The new trend is to have an Ethernet interface, though I believe the carrier is actually using MPLS to transport those packets. And the cable companies only use cable modems up to a certain point, then they are offering an Ethernet handoff, and I don't know what protocol they run behind it. In more rural areas, satellite Internet access is common, but it's not listed either. The form can't list all the possible transport protocols, but they seem to have missed some biggies. There will always be a need for an "Internet access: Other" item.

Third problem: not all applicants buy the local loop circuit with their Internet access. At that point, it doesn't really matter what technology is being used for the circuit (which is a Telecommunications Service).

Why does PIA want to know what the transport technology is, anyway? It's not like the information is reliable. Most of my clients who have a a 1.5 Mbps ATM or frame relay circuit think it's a T-1 (and the telecom salespeople often call it that), and I wouldn't know what to call an MPLS circuit with an Ethernet handoff, even if MPLS or Ethernet were an option. And few of the PIA people I've talked to have a good grip on this. Who would want the information if it were reliable? Does the FCC want to know how many applicants are using which protocols?

Finally, one of the options on the dropdown is "T-2," which is a 6 Mbps circuit that I've never heard of anyone using (apparently it died in the 1970s with the picturephone it supported). Typically, telcos will sell "bonded T-1s" until you get to 9 Mbps, at which point they switch to a "fractional T-3" (although they're just as likely to call it a "DS-3", which I think is more correct (which makes me wonder why you almost never hear anyone use "DS-1," which would be a more correct description of these circuits than "T-1," I think)).

And why not have boxes for "Bandwidth" and "Number of connections"? If PIA wants that info, the form should ask for it.

This week's horror

So last week I'm in the midst of gathering contracts for my clients so I can file their 471s, and I'm rejoicing over the final demise of the Two-Signature/Two-Date Rule (2s/2d), which I dubbed the Jason Rule because it just kept coming back after the FCC had apparently killed it, and what do I see a trailer for? Jason is back in the movie theaters.

Just a coincidence, I kept telling myself.

The Naughty Step

There seems to be a lot of grumbling around the office about the number of timeouts we're getting while putting in the Form 471. And I hear rumblings in the wider community. It's too bad; I had noticed timeouts disappearing. Has someone at USAC been inspired by the epic timeout battles on Supernanny? At least with Supernanny, when the kid gets "the Naughty Step" (so much quainter than "timeout"), s/he gets an explanation, and gets a hug at the end of the timeout. There are just not enough hugs in the E-Rate program.

Who's bleeding now?

Everyone seems pretty sanguine about the prospects for the E-Rate program under Obama and his pick for FCC Chairman, Julius Genachowski. So, of course, I'll take the contrarian view.

I'll start by talking about how the E-Rate managed to survive the Bush presidency and Rep. Joe Barton, who openly campaigned for it's demise. When Bush arrived in office, he wanted to end the E-Rate. What stopped him? I think it was the discovery that the funding never goes into the Treasury, so abolishing the E-Rate would not have freed up money for other priorities. And since the E-Rate brings money into every Congressional district in the country, Congress is reluctant to just kill it. Seeing that he couldn't outright kill it, Rep. Barton correctly reasoned that if he could get the E-Rate into the Treasury, he could "bleed it dry": once the E-Rate was just one of many competing priorities, it could slowly be cut and cut and cut. He was working on that until he lost his chairmanship when the Democrats took over the House.

Now here comes the Obama presidency, and everybody's happy with his new pick for FCC Chairman. They're both tech-friendly, and Genachowski even helped pen the original Report and Order. And Obama's Technology and Innovation Plan seems very pro-E-Rate. It sounds like he's going to expande the E-Rate: "Obama will recommit America to ensuring that our schools, libraries, households and hospitals have access to next generation broadband networks. He will also make sure that there are adequate training and other supplementary resources to allow every school, library and hospital to take full advantage of the broadband connectivity." (Will training become eligible? Or is it already?)

Nothin' but blue skies.

So what's that dark little cloud on the horizon?

Well, as more and more programs are paid from the Universal Service Fund, there will be pressure to take money from the E-Rate. How about those rollover funds we get every year? Why not take that unused funding and roll it into funding home broadband connections? I mean, every school has broadband, but most houses don't, so where is the priority? In the past, I worried that the ballooning High Cost program would take funding from the E-Rate. Now it looks like new programs will be competing for Universal Service funds.

How ironic if Rep. Barton's dream of killing the E-Rate comes true because of an expansion of the Universal Service Program.

I've got your maintenance right here

I just saw a request filed with the FCC from a service provider about Basic Maintenance of Internal Connections (BMIC). I don't know how the service provider would feel about me bandying their name about, so I'll just use "SP" to refer to them.

SP correctly identified a basic problem with BMIC: no one gets approval for these contracts in time to cover July, so now they have to decide whether to front the money in the hopes of later approval, or forego maintenance. But the remedy that SP proposes falls short.

SP suggests that applicants be allowed to make an annual payment, then if funding isn't approved, decide what to do at that point. But the math doesn't work: we're talking about 90% applicants, so if they're paying 10% of the contract amount, that will cover just over a month. As a result, in early August the applicant still won't have funding approval, and the 10% payment will have run out.

A single annual payment is half the solution. The other half is to allow applicants to pay get funding for the annual payment even if the contract does not run July 1-June30.

That would allow applicants to set up maintenance contracts to run June-May.

Let me give an example. Let's say the change I'm suggesting is implemented for 2009-2010. In that case, an applicant would sign a contract now to cover June 2010-May 2011, and put it on their 471. In that scenario, the applicant doesn't care if their BMIC FRNs are funded by July 1, as long as they get funded before the end of the funding year.

True, the first year would be ugly: applicants would have to have one contract to cover July 2009-May 2010, and another to cover June 2010-May 2011. But in subsequent years, it would be much better for applicants.